Improved document ranking in ontology-based document search engine using evidential reasoning

Abstract

This study presents a novel approach to document ranking in an ontology-based document search engine (ODSE) using evidential reasoning (ER). Firstly, a domain ontology model, used for query expansion, and a connection interface to an ODSE are developed. A multiple attribute decision making (MADM) tree model is proposed to organise expanded query terms. Then, an ER algorithm, based on the Dempster-Shafer theory, is used for evidence combination in the MADM tree model. The proposed approach is discussed in a generic frame for document ranking, which is evaluated using document queries in the domain of electrical substation fault diagnosis. The results show that the proposed approach provides a suitable solution to document ranking and the precision at the same recall levels for ODSE searches have been improved significantly with ER embedded, in comparison with a traditional keyword-matching search engine, an ODSE without ER and a non-randomness-based weighting model.